Added converters argument to some optimization algorithms (qiskit-c…

…ommunity/qiskit-aqua#1260)

* added converter argument to opt algorithms

* now penalty parameter is passed properly

* added docstrings for penalty

Co-authored-by: Manoel Marques <Manoel.Marques@ibm.com>
Co-authored-by: Cryoris <jules.gacon@googlemail.com>
Co-authored-by: Steve Wood <40241007+woodsp-ibm@users.noreply.github.com>

qiskit/optimization/algorithms/grover_optimizer.py

@@ -26,7 +26,7 @@
 from qiskit.circuit.library import QuadraticForm
 from .optimization_algorithm import (OptimizationResultStatus, OptimizationAlgorithm,
                                      OptimizationResult)
-from ..converters.quadratic_program_to_qubo import QuadraticProgramToQubo
+from ..converters.quadratic_program_to_qubo import QuadraticProgramToQubo, QuadraticProgramConverter
 from ..problems import Variable
 from ..problems.quadratic_program import QuadraticProgram
 
@@ -37,23 +37,35 @@ class GroverOptimizer(OptimizationAlgorithm):
     """Uses Grover Adaptive Search (GAS) to find the minimum of a QUBO function."""
 
     def __init__(self, num_value_qubits: int, num_iterations: int = 3,
-                 quantum_instance: Optional[Union[BaseBackend, QuantumInstance]] = None) -> None:
+                 quantum_instance: Optional[Union[BaseBackend, QuantumInstance]] = None,
+                 converters: Optional[Union[QuadraticProgramConverter,
+                                            List[QuadraticProgramConverter]]] = None,
+                 penalty: Optional[float] = None) -> None:
         """
         Args:
             num_value_qubits: The number of value qubits.
             num_iterations: The number of iterations the algorithm will search with
                 no improvement.
             quantum_instance: Instance of selected backend, defaults to Aer's statevector simulator.
+            converters: The converters to use for converting a problem into a different form.
+                By default, when None is specified, an internally created instance of
+                :class:`~qiskit.optimization.converters.QuadraticProgramToQubo` will be used.
+            penalty: The penalty factor used in the default
+                :class:`~qiskit.optimization.converters.QuadraticProgramToQubo` converter
+
+        Raises:
+            TypeError: When there one of converters is an invalid type.
         """
         self._num_value_qubits = num_value_qubits
         self._num_key_qubits = None
         self._n_iterations = num_iterations
         self._quantum_instance = None
-        self._qubo_converter = QuadraticProgramToQubo()
 
         if quantum_instance is not None:
             self.quantum_instance = quantum_instance
 
+        self._converters = self._prepare_converters(converters, penalty)
+
     @property
     def quantum_instance(self) -> QuantumInstance:
         """The quantum instance to run the circuits.
@@ -142,7 +154,7 @@ def solve(self, problem: QuadraticProgram) -> OptimizationResult:
         self._verify_compatibility(problem)
 
         # convert problem to QUBO
-        problem_ = self._qubo_converter.convert(problem)
+        problem_ = self._convert(problem, self._converters)
         problem_init = deepcopy(problem_)
 
         # convert to minimization problem
@@ -154,7 +166,7 @@ def solve(self, problem: QuadraticProgram) -> OptimizationResult:
                 problem_.objective.linear[i] = -val
             for (i, j), val in problem_.objective.quadratic.to_dict().items():
                 problem_.objective.quadratic[i, j] = -val
-        self._num_key_qubits = len(problem_.objective.linear.to_array())
+        self._num_key_qubits = len(problem_.objective.linear.to_array())  # type: ignore
 
         # Variables for tracking the optimum.
         optimum_found = False
@@ -259,7 +271,7 @@ def solve(self, problem: QuadraticProgram) -> OptimizationResult:
                                     status=OptimizationResultStatus.SUCCESS)
 
         # cast binaries back to integers
-        result = self._qubo_converter.interpret(result)
+        result = self._interpret(result, self._converters)
 
         return GroverOptimizationResult(x=result.x, fval=result.fval, variables=result.variables,
                                         operation_counts=operation_count, n_input_qubits=n_key,

qiskit/optimization/algorithms/minimum_eigen_optimizer.py

@@ -19,7 +19,7 @@
 
 from .optimization_algorithm import (OptimizationResultStatus, OptimizationAlgorithm,
                                      OptimizationResult)
-from ..converters.quadratic_program_to_qubo import QuadraticProgramToQubo
+from ..converters.quadratic_program_to_qubo import QuadraticProgramToQubo, QuadraticProgramConverter
 from ..problems.quadratic_program import QuadraticProgram, Variable
 
 
@@ -101,8 +101,9 @@ class MinimumEigenOptimizer(OptimizationAlgorithm):
         result = optimizer.solve(problem)
     """
 
-    def __init__(self, min_eigen_solver: MinimumEigensolver, penalty: Optional[float] = None
-                 ) -> None:
+    def __init__(self, min_eigen_solver: MinimumEigensolver, penalty: Optional[float] = None,
+                 converters: Optional[Union[QuadraticProgramConverter,
+                                            List[QuadraticProgramConverter]]] = None) -> None:
         """
         This initializer takes the minimum eigen solver to be used to approximate the ground state
         of the resulting Hamiltonian as well as a optional penalty factor to scale penalty terms
@@ -112,10 +113,17 @@ def __init__(self, min_eigen_solver: MinimumEigensolver, penalty: Optional[float
         Args:
             min_eigen_solver: The eigen solver to find the ground state of the Hamiltonian.
             penalty: The penalty factor to be used, or ``None`` for applying a default logic.
+            converters: The converters to use for converting a problem into a different form.
+                By default, when None is specified, an internally created instance of
+                :class:`~qiskit.optimization.converters.QuadraticProgramToQubo` will be used.
+
+        Raises:
+            TypeError: When there one of converters is an invalid type.
         """
         self._min_eigen_solver = min_eigen_solver
         self._penalty = penalty
-        self._qubo_converter = QuadraticProgramToQubo()
+
+        self._converters = self._prepare_converters(converters, penalty)
 
     def get_compatibility_msg(self, problem: QuadraticProgram) -> str:
         """Checks whether a given problem can be solved with this optimizer.
@@ -158,7 +166,7 @@ def solve(self, problem: QuadraticProgram) -> MinimumEigenOptimizationResult:
         self._verify_compatibility(problem)
 
         # convert problem to QUBO
-        problem_ = self._qubo_converter.convert(problem)
+        problem_ = self._convert(problem, self._converters)
 
         # construct operator and offset
         operator, offset = problem_.to_ising()
@@ -191,7 +199,8 @@ def solve(self, problem: QuadraticProgram) -> MinimumEigenOptimizationResult:
         # translate result back to integers
         result = OptimizationResult(x=x, fval=fval, variables=problem_.variables,
                                     status=OptimizationResultStatus.SUCCESS)
-        result = self._qubo_converter.interpret(result)
+
+        result = self._interpret(result, self._converters)
 
         return MinimumEigenOptimizationResult(x=result.x, fval=result.fval,
                                               variables=result.variables,

qiskit/optimization/algorithms/optimization_algorithm.py

@@ -20,6 +20,8 @@
 
 from .. import QiskitOptimizationError
 from ..problems.quadratic_program import QuadraticProgram, Variable
+from ..converters.quadratic_program_to_qubo import (QuadraticProgramToQubo,
+                                                    QuadraticProgramConverter)
 
 
 class OptimizationResultStatus(Enum):
@@ -35,85 +37,6 @@ class OptimizationResultStatus(Enum):
     """the optimization algorithm obtained an infeasible solution."""
 
 
-class OptimizationAlgorithm(ABC):
-    """An abstract class for optimization algorithms in Qiskit's optimization module."""
-
-    @abstractmethod
-    def get_compatibility_msg(self, problem: QuadraticProgram) -> str:
-        """Checks whether a given problem can be solved with the optimizer implementing this method.
-
-        Args:
-            problem: The optimization problem to check compatibility.
-
-        Returns:
-            Returns the incompatibility message. If the message is empty no issues were found.
-        """
-
-    def is_compatible(self, problem: QuadraticProgram) -> bool:
-        """Checks whether a given problem can be solved with the optimizer implementing this method.
-
-        Args:
-            problem: The optimization problem to check compatibility.
-
-        Returns:
-            Returns True if the problem is compatible, False otherwise.
-        """
-        return len(self.get_compatibility_msg(problem)) == 0
-
-    @abstractmethod
-    def solve(self, problem: QuadraticProgram) -> 'OptimizationResult':
-        """Tries to solves the given problem using the optimizer.
-
-        Runs the optimizer to try to solve the optimization problem.
-
-        Args:
-            problem: The problem to be solved.
-
-        Returns:
-            The result of the optimizer applied to the problem.
-
-        Raises:
-            QiskitOptimizationError: If the problem is incompatible with the optimizer.
-        """
-        raise NotImplementedError
-
-    def _verify_compatibility(self, problem: QuadraticProgram) -> None:
-        """Verifies that the problem is suitable for this optimizer. If the problem is not
-        compatible then an exception is raised. This method is for convenience for concrete
-        optimizers and is not intended to be used by end user.
-
-        Args:
-            problem: Problem to verify.
-
-        Returns:
-            None
-
-        Raises:
-            QiskitOptimizationError: If the problem is incompatible with the optimizer.
-
-        """
-        # check compatibility and raise exception if incompatible
-        msg = self.get_compatibility_msg(problem)
-        if msg:
-            raise QiskitOptimizationError('Incompatible problem: {}'.format(msg))
-
-    def _get_feasibility_status(self, problem: QuadraticProgram,
-                                x: Union[List[float], np.ndarray]) -> OptimizationResultStatus:
-        """Returns whether the input result is feasible or not for the given problem.
-
-        Args:
-            problem: Problem to verify.
-            x: the input result list.
-
-        Returns:
-            The status of the result.
-        """
-        is_feasible = problem.is_feasible(x)
-
-        return OptimizationResultStatus.SUCCESS if is_feasible \
-            else OptimizationResultStatus.INFEASIBLE
-
-
 class OptimizationResult:
     """A base class for optimization results.
 
@@ -279,3 +202,156 @@ def variable_names(self) -> List[str]:
             The list of variable names of the optimization problem.
         """
         return self._variable_names
+
+
+class OptimizationAlgorithm(ABC):
+    """An abstract class for optimization algorithms in Qiskit's optimization module."""
+
+    @abstractmethod
+    def get_compatibility_msg(self, problem: QuadraticProgram) -> str:
+        """Checks whether a given problem can be solved with the optimizer implementing this method.
+
+        Args:
+            problem: The optimization problem to check compatibility.
+
+        Returns:
+            Returns the incompatibility message. If the message is empty no issues were found.
+        """
+
+    def is_compatible(self, problem: QuadraticProgram) -> bool:
+        """Checks whether a given problem can be solved with the optimizer implementing this method.
+
+        Args:
+            problem: The optimization problem to check compatibility.
+
+        Returns:
+            Returns True if the problem is compatible, False otherwise.
+        """
+        return len(self.get_compatibility_msg(problem)) == 0
+
+    @abstractmethod
+    def solve(self, problem: QuadraticProgram) -> 'OptimizationResult':
+        """Tries to solves the given problem using the optimizer.
+
+        Runs the optimizer to try to solve the optimization problem.
+
+        Args:
+            problem: The problem to be solved.
+
+        Returns:
+            The result of the optimizer applied to the problem.
+
+        Raises:
+            QiskitOptimizationError: If the problem is incompatible with the optimizer.
+        """
+        raise NotImplementedError
+
+    def _verify_compatibility(self, problem: QuadraticProgram) -> None:
+        """Verifies that the problem is suitable for this optimizer. If the problem is not
+        compatible then an exception is raised. This method is for convenience for concrete
+        optimizers and is not intended to be used by end user.
+
+        Args:
+            problem: Problem to verify.
+
+        Returns:
+            None
+
+        Raises:
+            QiskitOptimizationError: If the problem is incompatible with the optimizer.
+
+        """
+        # check compatibility and raise exception if incompatible
+        msg = self.get_compatibility_msg(problem)
+        if msg:
+            raise QiskitOptimizationError('Incompatible problem: {}'.format(msg))
+
+    def _get_feasibility_status(self, problem: QuadraticProgram,
+                                x: Union[List[float], np.ndarray]) -> OptimizationResultStatus:
+        """Returns whether the input result is feasible or not for the given problem.
+
+        Args:
+            problem: Problem to verify.
+            x: the input result list.
+
+        Returns:
+            The status of the result.
+        """
+        is_feasible = problem.is_feasible(x)
+
+        return OptimizationResultStatus.SUCCESS if is_feasible \
+            else OptimizationResultStatus.INFEASIBLE
+
+    def _prepare_converters(self, converters: Optional[Union[QuadraticProgramConverter,
+                                                             List[QuadraticProgramConverter]]],
+                            penalty: Optional[float] = None) -> List[QuadraticProgramConverter]:
+        """Prepare a list of converters from the input.
+
+        Args:
+            converters: The converters to use for converting a problem into a different form.
+                By default, when None is specified, an internally created instance of
+                :class:`~qiskit.optimization.converters.QuadraticProgramToQubo` will be used.
+            penalty: The penalty factor used in the default
+                :class:`~qiskit.optimization.converters.QuadraticProgramToQubo` converter
+
+        Returns:
+            The list of converters.
+
+        Raises:
+            TypeError: When the converters include those that are not
+            :class:`~qiskit.optimization.converters.QuadraticProgramConverter type.
+        """
+        converters_ = []  # type: List[QuadraticProgramConverter]
+        if converters is None:
+            converters_ = [QuadraticProgramToQubo(penalty=penalty)]
+        elif isinstance(converters, QuadraticProgramConverter):
+            converters_ = [converters]
+        elif isinstance(converters, list) and \
+                all(isinstance(converter, QuadraticProgramConverter) for converter in converters):
+            converters_ = converters
+        else:
+            raise TypeError('`converters` must all be of the QuadraticProgramConverter type')
+
+        return converters_
+
+    def _convert(self, problem: QuadraticProgram,
+                 converters: Union[QuadraticProgramConverter,
+                                   List[QuadraticProgramConverter]]) -> QuadraticProgram:
+        """Convert the problem with the converters
+
+        Args:
+            problem: The problem to be solved
+            converters: The converters to use for converting a problem into a different form.
+
+        Returns:
+            The problem converted by the converters.
+        """
+        problem_ = problem
+
+        if not isinstance(converters, list):
+            converters = [converters]
+
+        for converter in converters:
+            problem_ = converter.convert(problem_)
+
+        return problem_
+
+    def _interpret(self, result: OptimizationResult,
+                   converters: Union[QuadraticProgramConverter,
+                                     List[QuadraticProgramConverter]]) -> OptimizationResult:
+        """Convert back the result of the converted problem to the result of the original problem.
+
+        Args:
+            result: The result of the converted problem.
+            converters: The converters to use for converting back the result of the problem
+                to the result of the original problem.
+
+        Returns:
+            The result of the original problem.
+        """
+        if not isinstance(converters, list):
+            converters = [converters]
+
+        for converter in converters[::-1]:
+            result = converter.interpret(result)
+        return result